Multi-turn agent benchmarking with ACP — run any agent, any model, any provider.
benchflow is a LLM eval & observability project. It focuses on evaluating and benchmarking the performance of multi-turn AI agents across different providers. benchflow is an open-source project aimed at AI researchers and developers. benchflow is open source under the Apache-2.0 license. benchflow is available on the web and the command line.
Behind benchflow is benchflow-ai, and it first shipped in 2025. Development happens publicly on GitHub with 255 stars and 1k commits in the last 90 days. Among its 4 catalogued features are agent benchmarking, multi-provider support, and multi-turn evaluation.
What PulseGate has recorded for this listing
Closest matches by what these projects do